for review: printing variable names and values in the tracer

[Erwan JAHIER]

Fergus Henderson wrote:
> 
> On 01-Feb-1998, Zoltan Somogyi <zs at cs.mu.OZ.AU> wrote:
> >
> > runtime/mercury_trace.[ch]:
> >       Change the prompt to a less intrusive "> ".
> 
> Certainly the previous prompt was pretty ugly, but I think "> " is too
> ambiguous -- since the output of both the debugger and the program
> being debugged go to the same stream, it is too easy to confuse them.
> I suggest one of the following (in approximate order of preference):
> 
>         - "mrdb> "      (mercury runtime debugger)
>         - "Mercury Runtime> "
>         - "MR> "
> 
> 
> Note that I use the name "mrdb" rather than "mdb" becaused
> "mdb" should probably be reserved for the bytecode debugger.
> (Of course, perhaps we should eventually integrate the two...)
> 

Note that in the "first version of Opium for Mercury"(1) that I'am
currently implementing, the queries and the display of thoses queries
are not in the same process than the "program being traced": so there
will be no stream confusion.

Morever, (Mireille and) I would not call the current Mercury tracer a
debugger but just a tracer ! ;-)

(1) e.g. Prolog top level loop + a primitive to move forward in the
trace history + a primitive to retrieve information about the current
event + stuff to communicate with the traced program.
-- 
R1.